Anti-cancer drug discovery and development is moving towards a more rational and targeted approach based on our current understanding of the molecular pathogenesis of a variety of human cancers. The application of these new molecularly targeted agents to the treatment of childhood cancers is a focus of this project. The ras family of G-proteins play an important role in the transduction of signals that trigger cell proliferation, and mutations in ras genes are found in 30% of all human cancers. Ras proteins undergo post-translational farnesylation, which is required for activity of wild-type and mutant ras proteins, and this step can be inhibited by farnesyltransferase inhibitors, such as R115777. Patients with neurofibromatosis type 1 (NF1) have an increased risk of developing tumors of the central and peripheral nervous system, with no standard treatment options, other than surgery available. Neurofibromin, which is the product of the NF1 gene, contains a domain with significant homology to ras GTPase-activating proteins. Decreased levels of neurofibromin have been shown to be associated with a constitutively activated ras-GTP status. The evaluation of R115777 in children with refractory solid tumors and neurofibromatosis type I (NF1) is therefore a rational choice. A phase I trial of R115777 for children with these tumors was recently completed, and based on the results of this phase I trial, a multi-institutional, randomized, double-blinded, placebo-controlled, cross-over phase II trial of R115777 for patients with NF1 and progressive plexiform neurofibromas has been developed and will soon open. The endpoint of this trial will be timed to disease progression. In addition, based on a 30% response rate to R115777 in adults with refractory leukemias, we will also soon open a phase I trial of R115777 for children with refractory leukemias. A series of pharmacodynamic studies evaluating the effect of R115777 will be included in the NF1 and leukemia trials. The clinical development of antimetabolites, such as raltitrexed, and agents that modualte the effects of antimetabolites, such as the recombinant bacterial enzyme, carboxypeptidase-G2 (CPDG2), is also being studied. CPDG2 hydrolyzes methotexate (MTX) to inactive metabolites. We have extensively evaluated the use of CPDG2 as a rescue agent for patients with high-dose MTX (HDMTX) induced renal dysfunction. CPDG2 provides an alternative route of elimination for MTX and plasma MTX concentrations decline by >95% within minutes in all patients. We have studied the pharmacokinetics of 2,4-diamino-N10-methylpteroic acid (DAMPA), the product of MTX hydrolysis by CPDG2. Three DAMPA metabolites have been identified and account for the more rapid elimination of DAMPA compared to MTX in patients who receive CPDG2 for HDMTX-induced renal dysfunction. A New Drug Application for the use of CPDG2 in HDMTX induced renal dysfunction will be filed based on these data. We are also evaluating the potential benefit of intrathecal (IT) CPDG2 administration to patients who receive accidental IT MTX overdoses. To date three patients with ALL who had received accidental IT MTX overdoses from 190 mg to 600 mg were entered on this multi-institution protocol. All three patients tolerated IT CPDG2 administration well, experienced a dramatic decrease in cerebrospinal fluid MTX concentrations, and completely recovered from MTX-associated toxicities.